How Dell successfully redesigned the Alienware website: Q&A with Julie Vittengl, Dell’s Senior Taxonomist and User Experience Researcher

| May 15, 2017
Sign up to get weekly resources, and receive your FREE bonus eBook.
Thank you!

Get ready for some great content coming to your inbox from the team at UserTesting!

In a recent webinar, Julie Vittengl, Senior Taxonomist and User Experience Researcher at Dell shares how her team helped ensure that the enterprise website redesign was successful. She discusses how the Digital Customer Experience research team is organized and how they get insights into how customers navigate the site.

We had a great Q&A session with Julie and included some of our favorite questions below, or you can watch the full webinar here. Enjoy!


How do you quantify the importance of qualitative research to stakeholders who own budgets?

I think a lot of our qualitative research helps inform some of the actual quantitative research and testing that we then follow up with on and on The things that we learn by talking to our customers that they tell us are important, we can actually spin up tests through our test and target organization and put them on, or, on in this instance, and get real data for how those changes, suggested by the qualitative data, will impact performance, and revenue, and all of those other important metrics. For us at Dell, it’s quantitative informing how we get qualitative and then showing actual data based on that for the website.

How long did the process for research and redesign take?

This was maybe a six to eight-month process from start to finish. My piece, which was taxonomy research, took place over about three months, but that was run concurrently with other forms of research and redesign was going on at the same time.

Sometimes things like visuals and templates don’t require the taxonomy research to be completed. It’s not really a waterfall approach. There’s a lot of different things going on at the same time where taxonomy research can happen at the same time as the content audit.

Was this a one-time effort or is it ongoing?

This was a one-time effort to redesign, get it off of its old platform and onto a new platform and create new templates for it. These types of projects that are long-term don’t happen very often, but there’s always different types of projects like this going on.

Once we get one part of the site up to current standards, then we move on to another part of the site. Then, at some point, we’ll go back and we’ll look at again. Probably not for a few years. There may be other parts of the site that get optimized after the fact and might be smaller one-off projects rather than a total site redesign.

What kind of tools do you use for the taxonomy testing?

Optimal Workshop is a suite of tools. The card sorting tool, like I said, is a digital version of the old analog card sorts where you could give a room full of people some index cards and ask them to sort them into groups.

If you’re using Treejack, which is the taxonomy testing tool, you need to already have a taxonomy. You already have to have an information architecture already defined to test in this case. What you do there is you give people scenarios. You say, “Where would you go to find X?” Then people click through your tree and try and find that thing you’ve asked them to find. That gives you the click task, the success rate, the findability, and the directness.

The Chalkmark tool really just validates that first impression of either your website, or a specific tool, or whatever it is you want somebody to test. You ask them something like, “Okay, you’re looking for X. Where would you first click?” You don’t get click tasks or anything like that. All you get is the heat map that shows where on that page people expect to be able to go to find an answer.

Are you also testing for mobile phones, or is it just for desktop?

The tests can be taken on any device. Typically we’re testing desktop. Right now at, we do not have a different hierarchy for our mobile experience than we do for our online experience. I know there are some online retailers that do this. They’ve (what they consider) optimized their taxonomy. Perhaps presenting a smaller subset of their categories to their users. Our navigation does look different, depending on your experience. What you see on on a laptop or desktop for the masthead navigation is different than what you would see on a mobile device, but we don’t test one hierarchy versus another. It’s definitely something to think about, but it’s not anything that we’ve done as of this time.

Is your team agile? How does it typically work while you’re doing a three or six-month redesign?

They are agile. Especially within the research and testing organization, but I would say that the Digital Customer Experience team as a whole operates with a fairly agile mentality. We work as quickly as we can.

We put tests and prototypes in front of users as often as we can and we get that feedback immediately to our design team and our architects to make improvements, and to update the visuals, and to change the functionality. Then we get those redesigns right back out there in front of the different audiences that we’re testing with to see if we’re finding improvements or if we need to redesign again before we submit final comps to our development teams. Then the development team obviously is on a different cycle. DCX, Digital Customer Experience, does not do the actual development for the website. They do the design work, all of the templates and visual design.

How does a taxonomist impact the research process?

When I was first hired as a taxonomist at Dell, I was primarily involved in governance around how taxonomy is presented on Research was something that I started doing a little bit later in my job as it became apparent that every time there was a project for the website that the Digital Customer Experience was working on, taxonomy impacted that somehow.

I am a cross-department resource. I work with various different teams that are focused on different parts of the website as a resource for them in their design projects. I do like to be involved in every project they’re working on just to understand if there is a taxonomy impact. Sometimes there’s not, but often times there is. I think it’s important, as you look at different parts of the website and do redesigns, that you try and understand how your content is organized and how taxonomy might impact that design.

Does taxonomy differ for different demographics?

Yeah, it definitely does. Especially if it’s a consumer versus a business person. On we have one taxonomy that we present to consumers when they segmented themselves and another taxonomy that we’ve presented to our business customers when they’ve segmented themselves. I think we’d really love to get to a point where it’s totally personalized based on how people interact with the site. We’re not quite there yet.

For example, consumers will see products like gaming, and cameras, and different types of accessories that our business customers won’t see. There is one what I’ll call, “master taxonomy.” It’s all the products that Dell might sell, but depending on how somebody interacts with the website and how they self-segment, they’re presented with a subset of that taxonomy. They’re very similar. We have laptops for consumers and we have laptops for business people. Then once you get into more the electronics and accessories, on the consumer side you’re going to see a lot more products for the consumers.

Then on the business side, you’ll see things like servers, and storage equipment, and networking devices. Those things are not presented to consumers. We don’t expect consumers to buy things like servers. It’s not to say that they couldn’t, it’s just that that’s not part of the primary taxonomy experience we give them.

How do you advocate for story prototyping before the visual designing begins?

When I first came into this role, often times I got brought into a project after it already started and some of the visual work was done. Of course, you want your design to fit the taxonomy, not the taxonomy to fit the design. If somebody builds a page or a template and it says we can only have five categories, but you’ve got seven things, what are you going to do with those other things? You can’t not show them to the user because the template won’t let you.

I think in general, now that I’ve been in this role for three and a half years, I think it’s just part of the process now, that I’m consulted at the start, because people want to know, like the designers and the content strategists, what the taxonomy needs are for this project.

I start working as soon as we get the go-ahead that a project’s on the roadmap to take a look at the taxonomy, understand what its current state looks like, understand what the findability is. As it stands today, I’m almost never working with a brand new taxonomy. I’m always looking at some kind of redesign of something that already exists. I can get started right away doing current-state findability, understanding where our users are stumbling today, then start working on the iterations to understand where we need to take the taxonomy and make sure that the teams have that feedback right away.

What were the critical success factors for taxonomy? Did you have a goal in mind before you started?

For taxonomy, I don’t think there were any specific goals. I think it was more understanding whether or not the current organization worked for users or if we needed something different.

Obviously, revenue per visit is huge for Dell. We’re an eCommerce website so one of the main KPIs is understanding whether or not we’re making more money every time somebody visits our site. The taxonomy redesign, by placing, for example, the Alienware Alpha in more than one category—it sits with the desktops and it sits with the consoles—gives additional visibility into that product.

I think having a high success rate for findability helps people find products more quickly and then ultimately drives that revenue. If people could find the product quickly, then they’ll put it in their cart more quickly, they’ll check out, and buy it.